1. Failure Mode and Effects Analysis FMEA
Everything that can fail, shall fail!
Failure Mode and Effects Analysis FMEA
Introduction to FMEA
Failure Mode and Effects Analysis (FMEA) is an important reliability design tool that has been gaining in popularity.

2. Failure Mode and Effects Analysis Definition
A bottoms-up, iterative approach for analyzing a design of a product or process in order to determine
what could wrong
how badly it might go wrong
and what needs to be done to prevent it
It is a systematic bottoms-up approach for identifying failure modes, their causes and effects, and their elimination during conceptual design.

3. Alternate Definition
Failure Mode and Effects Analysis (FMEA for short) is a systematic way to recognize and evaluate the potential failures of a product or process. It provides a formal mental discipline for eliminating or reducing the risks of product failure. It also serves as a living document, providing a method of organizing and tracking concerns and changes through product development and launch.
A formal definition.

4. Yet Another Definition
Any formal, structured activity which is applied in developing something new to assure that as many potential problems as are reasonably possible to predict have considered, analyzed, and their causes remedied before the item under development reaches the hands of the end user.
Applicable to
product development
idea development
organization development
process development
software development
FMEA can be applied to either a product or a process.

5. Failure Mode and Effects Analysis Objective
…to identify early in the product or the manufacturing process design all manner of failures so they can be eliminated or minimized through design correction at the earliest possible time.
The proper use of this technique allows for the early identification of reliability problem areas.

6. Failure Mode, Effects, and Criticality Analysis (FMECA)
What’s a FMECA?
A more expanded version of FMEA includes a determination of the criticality or severity of a particular failure mode.

7. Benefits Improved product or process functionality
Verify design integrity
Provide rationale for change
Reduced warranty and replacement costs
Reduction in day-to-day manufacturing problems and costs
Improved safety of products and processes
The FMEA process provides for a critical look at the design by a number of knowledgeable people. Collectively, this study group is often able to identify a number of areas for improvement.

8. Introduction to FMEA Background Basic Concepts
Product versus Process Application
Overview of Methodology
Related Concepts
The remainder of the discussion will follow this outline.

9. Background 1949 - US military
Military Procedure MIL-P-1629 (procedures for performing a FMEA
used as reliability evaluation technique
1960’s - Used in the by the aerospace industry and NASA during the Apollo program
1988 – ISO 9000 business management standards
required organizations to develop quality systems
QS 9000 developed by Chrysler, Ford and GM
compliant automotive suppliers shall utilize FMEA
1993 – Automotive Industry Action Group (AIAG) and American Society for Quality Control (ASQC)
Society of Automotive Engineers (SAE) procedure SAE J-1739
Provides general guidelines for performing a FMEA
Some history.

10. Basic Concept begin at the lowest level of the system
identify potential failure modes
assess their effect and causes
prioritize based upon effect
through redesign
eliminate the failure
or mitigate its effect
FMEA is a bottoms-up approach for identifying failures modes and their causes, prioritizing their impacts, and then seeking ways to either eliminate or reduce their occurrence.

11. Basic Concept - Example
component – computer monitor
part – capacitor
identify two failure modes
fail “open”
effect are wavy lines appearing on monitor
fail “short”
effect is the monitor going blank
prioritize – short more critical than open
determine cause of failure mode
underrated capacitor
investigate ways of eliminating failure
resize capacitor
A simple example of the overall approach.

12. More Basic Concepts Team effort
5 to 7 members
team lead engineer
representation from design, assembly, manufacturing, materials, quality, and suppliers
Usually done near the end of the product or process design phase
Analysis should continue throughout the product development cycle
Should be a living document that is updated as design changes and new information becomes available
Teams are formed to conduct a FMEA study. The team is responsible for conducting and documenting the analysis and then making recommendations for implementation.

13. Product versus Process
Product or Design FMEA.
What could go wrong with a product while in service as a result of a weakness in design.
Product design deficiencies
Process FMEA.
What could go wrong with a product during manufacture or while in service as a result of non-compliance to specification or design.
Manufacturing or assembly deficiencies
Focus on process failures and how they cause bad quality products to be produced
The technique can be applied to a process as well as a product. For example, a FMEA may be conducted to evaluate a manufacturing process for a new product.

14. Product (Design) FMEA
Assumes manufacturing and assembly will produce to design specifications.
Does not need to include failure modes resulting from manufacturing and assembly.
Does not rely on process controls to overcome design weaknesses.
Does consider technical and physical limitations of the manufacturing and assembly process.
A product FMEA will usually take the production process as a given. That is, it will assume that manufacturing will produce to specifications.

15. Process FMEA Assumes the product meets the intent of the design.
Does not need to include failure modes originating from the design.
assumes a design FMEA covers these failures
Usually originates from a flow chart of the process
A process FMEA will not normally address product design failure modes.

16. The FMEA Team FMEA Team manufacturing or process engineer quality
operations
facilitator
maintenance
Typical team members for a process FMEA. The facilitator is knowledgeable in conducting a FMEA but may not have in-depth process or product knowledge.
product
engineer

17. Affected Functional Areas
reliability
vendors
customers.
downstream engineering functions
downstream manufacturing functions
end users
service functions,
recycling or reuse functions
design
materials
manufacturing
assembly
packaging
shipping
service
recycling
quality
The scope of an FMEA may be quite large and encompass the entire production environment.

18. Methodology System or Process Definition
Determination of Failure Modes
Determination of Cause
Assessment of Effect
Estimation of Probability of Occurrence (O)
Estimation of Detecting a Defect (D)
Classification of Severity (S)
Computation of Criticality (Risk Priority Number)
RPN = (S) x (O) x (D)
Determination of Corrective Action
Nine easy steps to conducting a FMEA. Step 5, 6, and 7 requires some level of quantification in to prioritize failure modes in step 8.

19. FMEA Flow Diagram 2. Identify Failure Modes 5. Estimate occurrence
8. Compute
RPN
Prioritize
1. Define
Process
3.Establish
Cause
6. Determine
Detection
The nine easy steps shown graphically.
4. Assess
Effect
7. Assign
Severity
9. Take
Corrective
Action

20. FMEA Worksheet Component or Process Failure Mode Failure Cause
Failure Effect
Correction
CRT Picture tube
Bad pixels
excessive heat
picture degraded
larger fan
dropping or bumping
improve packaging
Cabling to unit
broken or frayed
fatigue, heat
will not conduct
higher grade wire
internal short
heat, brittle insulation
shock, damage to unit
Documentation of the FMEA may be accomplished through the use of specialized software or through the use of a spreadsheet.

21. Methodology - Example
Perform a FMEA analysis for the process of installing a roof.
A simple example.

22. An example of what a spreadsheet might look like.

23. Step 1. Product / Process Definition
Describe product and its design or the process and its operations
Identify the purpose or function of each component or each operation
Use functional diagrams, design drawings, flow charts and other graphical techniques
Include each significant element that is likely to fail
Begin by defining the product and its design or the process and its operations.

24. Step 2. Determination of Failure Modes
A failure mode is the manner in which a process could potentially fail to meet the process requirement or the design intent.
It is a statement of non-performance or a non-conformance to a design specification.
Questions to be answered include:
how can the process/part fail to meet specs
regardless of the specs, what would customer find objectionable?
Each failure mode would be an entry (row) on the spreadsheet.

25. Examples of Failure Modes
ruptures
fractures or cracks
short or open circuits
deformation
contamination
loss of power
buckling
Failure modes may initially be determined through a brainstorming session by the FMEA team.

26. Step 3. Determination of Cause
Identify how the failure could occur
State in terms of something that can be corrected
Attempt to establish an exhaustive list
Further analysis may be required to isolate cause (e.g. a design of experiments)
Through knowledge of the product and its use or the process and its operations or through engineering analysis, failure causes are identified.

27. Example Causes improper tolerances or alignment operator error
part missing
cyclical fatigue
poor workmanship
defective parts from supplier
maintenance induced
aging and wear-out
excessive environmental conditions
For each failure mode, one or more causes are identified.

28. Failure Mode and Cause
By identifying the cause, the solution becomes more apparent.

29. Step 4. Assessment of Effect
Assess the effect of the failure mode on the customer
Customer may be next operation, subsequent operations, the end-user, or the seller
Answer the question what might the customer observe or experience.
Not all failures have the same effect particularly upon the customer. The intent is to find the failures that are most significant and important to the customer. A blemish on the cabinet housing of a stereo player is not as critical as a poor electrical connection.

30. Assessment of Effect - examples
In assessing the effect, determine what the final outcome of the failure will be.

31. Step 5. Estimation of Probability of Occurrence (O)
Occurrence refers to how frequently the specific failure mode will be observed.
Estimated on a scale from “1” to “10”
Statistical analysis may be used if historical data is available
Otherwise estimated subjectively
To prioritize failures, estimates of their (relative) probabilities of occurrences must be made.

32. Step 6. Estimation of Detecting a Defect (D)
The probability that the current process controls will detect the failure mode before the part or component leaves the process.
Assume failure has occurred, and then assess the likelihood that the product will continue to its next stage.
Rank on scale of “1” (almost certain to detect) to “10” (no way of detecting failure)
Some failures will be detected by processes already in place such as quality control and inspection. If a failure is detected before the product gets to the customer, then its significance is not as great.

33. Step 7. Classification of Severity (S)
An assessment of the seriousness of the effect of the failure mode on the customer
Estimated on a scale of “1” to “10.”
Assessed against
safety; i.e. injury or death
extent of damage
or amount of economic loss
The severity of the failure if observed by the customer is determined.

34. Step 8. Computation of Criticality
Risk Priority Number (RPN)
Product of Severity (S), Probability of Occurrence (O), and Detecting a Defect (D)
RPN = (S) x (O) x (D)
Range is 1 to 1000 with the higher the number, the more critical the failure mode.
Rank order RPN from highest to lowest
A risk priority number is just one way of ranking the various failure modes. Those failure modes with the highest scores should be addressed first.

35. Step 9. Corrective Action
Removing the cause of the failure,
Decreasing the probability of occurrence, or
Increase the likelihood of detection, or
Reducing the severity of the failure.
There are four general steps that can be taken to mitigate a failure mode.

36. Related Concepts Quality Functional Deployment (QFD)
customer requirements
Total Quality Management (TQM)
Statistical Process Control (SPC)
detection
Design of Experiments (DOE)
root causes
Six Sigma
process improvement
Fault Tree Analysis (FTA)
On-going Reliability Testing (ORT)
Other product design tools that are closely associated with a FMEA.

37. Difficulties in Implementation
Time and resource constraints
Lack of understanding of the purpose of FMEA
Lack of training
Lack of management commitment
Dale and Shaw, 1990: “Failure Mode and Effects Analysis in the Motor Industry,” Quality and Reliability Engineering International.
A FMEA must have management support, be adequate resourced, and the team properly trained in its objectives and use.

38. Failure Mode and Effects Analysis - FMEA
Questions?